Two burial caves with arcosolia were excavated. The finds date from the Late Roman and Byzantine periods (fourth–sixth centuries CE). Also excavated was a habitation level containing pottery sherds and glass industrial debris (Gorin-Rosen, below).
 
Burial Caves
The two caves (F1, F3; Fig. 2) were hewn in the kurkar bedrock. Both were reached by way of an entrance leading from the southwest to the northeast, extending deep into the kurkar ridge. The two are arcosolia type caves that consisted of an opening, an entrance step, a burial chamber and three arcosolia, one in each wall.
 
Cave F1 (Figs. 3, 4). Two of steps built of partly dressed medium-sized stones and another step built of square marble slabs were discovered at the entrance to the cave. The steps led into a square burial chamber (L123; 2.60 × 2.65 m), whose floor and walls were plastered. The floor (L126) was treated with white plaster, and gray-white plaster was applied to the walls; remains of red paint and black geometric decorations still adorned them (Fig. 5). The plaster on the floor adjoined the built marble step and abutted the plaster on the walls. Three rock-hewn and plastered arcosolia (L111, L114, L115; average dimensions 0.9 × 2.3 m, max. height 1.6 m) were hewn in three of the walls. The burial chamber was discovered filled with layers of clay alluvium and pale sandy soil, from the apex of the arcosolia to the top of the arcosolia—evidence of it having been open for a long time. Clay soil that contained fragments of a calvaria, a femur and a mandible that exhibited developmental characteristics of an individual aged 14–16 was excavated from the elevation of the tombs down to the floor. The bones, which were scattered in the entrance area, seem to indicate that the tomb was robbed. The excavation of the three arcosolia yielded finds dating from the Late Roman and the Byzantine periods (fourth–sixth centuries CE), including pottery sherds, glass shards, bronze coins and bone fragments.
The pottery from the northern arcosolium (L111, L117) consisted mainly of bag-shaped jars (Fig. 6:3), from the fifth–sixth centuries CE, and a jar (Fig. 6:5), whose upper part was intentionally broken by means of a series of small drilled holes in the upper two-thirds of the vessel, at its point of maximum diameter. Other discoveries included two bronze coins, one possibly from the fourth century CE (IAA 157666) and the other a pentanummium of Justinian I (551–560 CE; Antioch mint; IAA 157665). The bone fragments from this tomb represent at least fourteen individuals, including two children aged 2–10 years and twelve adult individuals. Judging by the mandibles and the closure of the epiphyses on the femurs, it was possible to identify two individuals 18–25 years old, an individual at least thirty years old, an individual 25–40 years old and eight individuals above the age of twenty. Five of the deceased were identified as female and two as male.
The pottery from the eastern arcosolium (L114, L116) consisted mainly of bag-shaped jars. The bone repertoire from this tomb included fragments of a calvaria, teeth and postcranial bones. Some of the bones were anatomically articulated, attesting to a primary burial. The individual was placed in a general north–south direction, with his head in the north. The rest of the bones were found scattered, representing at least six individuals: two children, ages 2–5 and 5–10; a male over the age of 60; two females over the age of 19; and one individual of undetermined sex, 19 or older.
The excavation in the southern arcosolium (L115, L124) exposed mainly sherds of bag-shaped jars. The bones recovered from this tomb represent at least nine individuals, including three infants and toddlersone aged 0.5–2.0 years, and the others 2–5 years of age—as well as six adult individuals: one aged 25–40, one older than 50, and four older than 19. At least four females and two males were identified among the adult individuals.
 
Cave F3 (Figs. 7, 8). Two layers of white plaster (average thickness of each 1 cm), containing fragments of glass and charcoal, coated the entrance to the burial cave. The plaster extended to the side of the burial chamber and covered the one step that descended down to the center of the chamber’s floor. The chamber was square (L107; 2.2 × 2.3 m) and paved with a white mosaic floor (L121; average size of the tesserae 3 × 3 cm). The walls of the burial chamber were completely covered with white plaster, which was stained with a charcoal additive or some other organic matter. Three rock-hewn, plastered arcosolia (L108–L110; average dimensions 0.65 × 1.90 m, max. height c. 1.5 m) were exposed on three sides of the chamber.
The burial chamber was found filled with fragments of plaster and kurkar, a result of the cave’s ceiling having collapsed (L104). The ceramic finds collected from the layer of collapse included Iron Age vessels, such as a krater (Fig. 6:2), as well as sherds from the Persian, Roman and Byzantine periods, the latter including mainly bag-shaped jars. In addition, fragments of rectangular kurkar slabs (average size 0.40 × 0.65 m) were also found scattered in the chamber. It is possible that the fragments of the covering panels of the tombs in the entrance area indicate that the grave had been plundered. Light sandy soil (average thickness 0.15 m) covered the mosaic floor. The arcosolia yielded artifacts dating from the Roman and Byzantine periods (fourth–sixth centuries CE), including potsherds, glass, bronze coins, metal jewelry and bone fragments.
Light sandy soil and fragments of the covering panels were found in the northern arcosolium (L108, L119). Pottery sherds, mainly of jars, and fragments of a metal bracelet (Fig. 9:1) and a silver earring (Fig. 9:2) were found while excavating the eastern arcosolium (L109, L120). The bone fragments in that tomb represent at least two individuals, one aged 20–30 years and the other is 19 or older. At least one of the individuals was a female, although it could not be determined which one; the second individual’s gender could not be determined.
While excavating the southern arcosolium (L110, L118) pottery sherds, mainly of jars, and three bronze coins were found. One of the coins is of Constantine I (posthumous, 337–341 CE, Nicomedia mint, IAA 157667), the second is dated generally to the fourth century CE (IAA 157668) and the third is from 383–395 CE (IAA 157669). A metal bracelet (Fig. 9:3) was also found. The bone fragments from this tomb represent at least one male individual 19 years of age or older.
 
All of the finds recovered from Caves F1 and F3 date to the Late Roman and Byzantine periods. The burials are identical to those in the other burial caves excavated in the Kfar Shmaryahu cemetery, and greatly resembles those that were discovered in Rishpon, Tel Baruch, Tel Qasile, Tel Arshaf, Herzliya B and C and elsewhere (see e.g, Gorzalczany 2000; Tal 1995; Tal and Taxel 2015). The ceramic types that were found are identical to those unearthed in the burial caves at these sites, and attest to similar funerary practices. These include, for example, the bottom part of the bag-shaped jar from Cave F1, which was probably used in a secondary burial (Fig. 6:5), and parallels found in other cemeteries (Tal and Taxel 2015: Fig. 1.46:1–2; 51–52). Similar metal artifacts, such as bracelets and earring (Fig. 9), were also found in these cemeteries (Tal and Taxel 2015: Figs. 1.9:3, 4; 1.15:18; 1.21:12; 1.34:9–14), as well as glass finds (Gorin-Rosen, below).
The human bones represent at least thirty individuals from Cave F1 and three individuals from Cave F3. Despite its poor state of preservation, the sample of human bones included infants, children and adult individuals, spanning a broad range of age groups. The two sexes are equally represented. This demographic cross-section is typical of an ordinary civilian population, and according to the number of people buried in the two caves, it is assumed that these are family burial caves, as was customary during this period.
 
Habitation Level (F2; Fig. 3)
A habitation level (L105, L106; thickness c. 0.4 m; Fig. 3) that lay on the Kurkar bedrock was discerned slightly west of the opening to Burial Cave F1. The level contained mostly jars from the Late Roman and Byzantine periods (Fig. 6:4), as well as pottery sherds from the Hellenistic period, such as a bowl base (Fig. 6:1), and glass kiln debris (Gorin-Rosen, below).
 
Glass Finds
Yael Gorin-Rosen
 
Three whole glass bottles and six vessel fragments (Fig. 10:1–9) were found in Tombs F1 and F3. Remains of kiln debris from glass manufacturing and flakes of raw glass were also discovered. The whole vessels represent familiar types from burial assemblages that were exposed in the region and date from the Late Roman–Byzantine periods (fourth–sixth centuries CE). The industrial debris is from kilns used in the manufacturing of raw glass, such as those that operated in the region of Apollonia and its surroundings during the Byzantine and Early Islamic periods (Roll and Ayalon 1989; Gorin-Rosen 2000:54–55; Tal, Jackson-Tal and Freestone 2004; 2008).
 
Glassware. An intact bottle (Fig. 10:2), a whole bottle with a broken rim (Fig. 10:1) and a bottle that was restored from numerous fragments (Fig. 10:3) were found, as well as fragments of at least four additional vessels: three bottles (Fig. 10:4–7; base 7 and one of the rims may belong to one bottle) and a jar or jug with a broad rim (Fig. 10:8, 9; both fragments may belong to the same vessel). The bottles in Fig. 10:1–3 have similar features; they seem to have come from the same workshop and may even have been made by the same craftsman. They are characterized by a relatively broad, rounded rim that is attached to the body or is a continuation of the neck that extends from the body. The body of the vessel is small in comparison to the width and height of the neck. None of these vessels has a clearly defined neck. The base is pushed in, and it bears no pontil scar. Very similar vessels, which have been identified with the Samaritan population, were found in tombs in the Tel Baruch cemetery (Jackson-Tal 2015).
Bottle No. 1 is made of almost colorless glass, with a pale yellowish green tinge, and is covered with iridescent silvery weathering and decay. The bottle is small and squat, and has a wide rounded funnel-like rim attached directly to a spherical–piriform body. The wall of the vessel is very thin. Four shallow grooves on the broad part of ​​the body were made by applying slight inward pressure to the body. This decoration is mainly known on closed vessels from the Late Roman and Early Byzantine periods. For example, similar vessels known as ‘thumb-indented jars’ were discovered in graves at Tel Baruch (Jackson-Tal 2015:61, Pl. 1.51:3, 4, see numerous other references cited therein). A bottle with a very similar form, rim treatment and decoration was found in a tomb at Khirbat el-Ni‘ana, which was dated to the end of the Late Roman period and the beginning of the Byzantine period (fourth–fifth centuries CE; Gorin-Rosen and Katsnelson 2007:94–95, Fig. 9:3).
Bottle No. 2, which is intact and made of pale greenish blue glass, is covered with silvery iridescent weathering that caused opacity, and some sandy weathering. The bottle has an upright, rounded and uneven rim, with a protrusion beneath it, and it is somewhat flattened toward the neck, forming a shallow wavy fold. The neck of the bottle is broad and cylindrical and the body is spherical.
Bottle No. 3 consists of many fragments in different states of preservation. The bottle is made of a very pale shade of greenish blue glass, is covered with silvery iridescent weathering and is partially pitted. The rim of the bottle was rounded in fire and shaped as an elongated funnel extending from a broad cylindrical neck. The vessel has a spherical–piriform body. Bottles that are very similar to Nos. 2 and 3 were found at Tel Baruch, where it was explicitly stated that they had no pontil scar, just like the bottles from Kefar Shemaryahu (Jackson-Tal 2015:67, Fig. 1.54:2, 3, and see other references therein).
Bottle No. 4 is made of light bluish green glass, and is covered with silvery iridescent weathering. The bottle has a funnel-like rim similar to that of Bottle No. 3, which was haphazardly crafted.
Bottles Nos. 5 and 6 have a wide funnel-shaped rim that is decorated with thin, blue glass trails. The two bottles are very similar and probably belong to one vessel: No. 5 is made of bluish light green glass, covered with silvery iridescent weathering; No. 6 is made of greenish blue glass and is decorated with blue trails, of which the upper one is thicker than the rest it. Medium and large bottles with a wide funnel rim, which are decorated with trails. are well known from assemblages dating from the Late Roman and the Byzantine periods.
Base No. 7 is made of pale greenish glass, and covered with silver weathering. The base is very thin, pushed inwards and has no pontil scar. It may belong to one of the bottles or may have come from another vessel.
Vessel No. 8 is either a bottle, a jar or a jug. It is made of olive green glass, decorated with blue trails and covered with silver, golden and iridescent weathering. The vessel has a broad, unevenly folded-in rim and a short, wide neck. The beginning of the shoulder is wide and decorated with trails that are of darker shade than the vessel. The rim is broken on one side, and the vessel probably had a wide handle, making it a jug rather than a bottle or a jar. Jars generally have an open fold beneath the rim and are decorated in a variety of ways. A jar that has a rim with an open fold and is decorated with trails on the body was found in the cemetery at Tel Baruch (Jackson-Tal 2015:61, Fig. 1.51:2). The proportions of the rim and the shoulder are reminiscent of large bottles with a wide, unadorned funnel rim that were discovered in the cemetery at Khirbat el-Ni‘ana in assemblies dating to the end of the Late Roman period and Early Byzantine period (fourth–fifth centuries CE; Gorin-Rosen and Katsnelson 2007:100–101, Fig. 12:1, 2).
Base No. 9 was found together with Rim No. 8. In light of the unique hue of both fragments, it can be assumed that they come from the same vessel. The base, which is pushed inwards and concave, is broken in the center.
 
Kiln Waste (Fig. 11).Flakes of various sizes and shapes from chunks of raw glass (L101, L105, L106; Fig. 11:1–15); some had remains from the side of the kiln adhering to them. Most of the flakes are triangular in section. The glass is generally bluish green or pale green. Three fragments included glass that is yellowish brown in shade (Fig. 11:2, 8, 13). The bluish green hue is the predominant color among the finds, and it is the primary color found in the glass kilns that were excavated in Apollonia and Bet Eli‘ezer. Purposely made Brown-yellow glass was usually produced in separate kilns. The brown-yellow glass that was found on one fragment together with greenish blue glass was caused due to exposure to concentrations of iron or its location deep inside the kiln, far from the opening through which the hot air entered the installation, resulting in a reduction processes. Flakes of various sizes were collected and sold by weight for re-melting in kilns, where they were used in the manufacture of glassware and objects.
Also found were a fragment from the wall of a kiln that bore remains of green glass (Fig. 11:16) and fragments of glazed bricks (Fig. 11:17–19. Fragment No. 17 is large and has the remains of the end of a vitrified brick, on which four glazed sides are visible. The brick is made of small stones and red hamra soil that contained a large amount of organic matter that burned and left lacunae. The sides that were in contact with the inner part of the kiln were fired and vitrified in a greenish blue glaze. Fragment No. 18 belongs to a dark-green glazed brick. Vitrified mud-bricks were found in the walls of the kilns used to manufacture glass at Bet Eli‘ezer and in the debris found around the kilns (Gorin-Rosen 2000:52). Similar finds were discovered in Apollonia, such at the Ha-Horesh Street excavation, where a large amount of waste was found (Haddad et al. 2015), and at other sites where such kiln debris was found, including many sites excavated or surveyed in the vicinity of Herzliya, Kefar Shemaryahu, Raʽanana and Kefar Saba (Roll and Ayalon 1989). This area evidently served as a very large center for the production of raw glass, most of which was designated for export.
 
Past studies have linked the cemetery in Kefar Shemaryahu with the Samaritan population that lived in Apollonia during the Late Roman and Byzantine periods. The types of burial caves and the nature of various finds—pottery, metal, glass and jewelry—are in keeping with the results of excavations and past surveys in Kefar Shemaryahu and the surrounding area, regarding both the burials and the glass manufacturing industry. Therefore, the two burial caves should be considered an integral part of the Samaritan cemetery of Kefar Shemaryahu.